Book Description

The drainage performance of unbound granular material (UGM) is an important consideration in pavement design because the presence of excess moisture in UGM layers can eventually lead to premature failures. Recently, transportation agencies have been evaluating their granular base and subbase drainage and mechanical performance to ensure sufficient drainage capacity while maintaining adequate structural support to produce more sustainable pavement structures. Linking performance to UGM construction specification requires accurate characterization of UGM's mechanical and drainage performance and how physical and gradation parameters affect such performance. These evaluations led to an update of the specification requirements of UGM in many jurisdictions including Manitoba. In this research, constant head hydraulic conductivity, resilient modulus, permanent deformation, double ring infiltrometer, and falling weight deflectometer test methods were used in laboratory and field investigations. These tests were conducted to characterize the drainage and mechanical performance of ten UGM samples representing four different gradation bands. The laboratory test results were also used to investigate the reliability of the estimated hydraulic conductivity from the Moulton prediction model and from the Enhanced Integrated Climatic Model (EICM). Test results showed an improvement in resilient modulus and drainage quality for samples in gradation bands that specify larger maximum aggregate size and limited fines. A statistical analysis of the test results showed that D10 larger than 0.2mm and D60 larger than 8mm would guarantee higher stiffness and better drainage performance with a time-to drain of less than 5days for typical pavement cross-sections and a resilient modulus value exceeding 200MPa. The Moulton prediction model was found to provide a better approximation of hydraulic conductivity of the materials included in this study, while the EICM model was found to significantly overestimate the hydraulic conductivity for most of the samples.